“Nobody cares what you do in there.”
In 1923, Henry Frugès, a French industrialist, commissioned a rising star in the field of architecture to design some homes for the workers who labored to build packing cases in his factories in Lège and Pessac, near Bordeaux. The brightly hued houses were pieces of pure modernism, concrete cuboids shorn of decoration. They were starkly beautiful, in the way that modern architecture at its best can be. The architect’s name was Charles-Édouard Jeanneret-Gris. Today we know him as Le Corbusier.
Le Corbusier admired the energy of modern industry and scorned the messy rural vernacular of French peasant architecture. Where was the beauty in that? Le Corbusier loved straight lines, elegant curves, and smooth planes: “These things are beautiful because in the middle of the apparent incoherence of nature of the cities of men, they are places of geometry, a realm where practical mathematics reigns,” he once wrote. “And is not geometry pure joy?”
Reasonable people may disagree with that, and there was one group that did not care much for Le Corbusier’s geometry: the humble French workers who lived in his joyful cubes at Pessac, trudging home each night after a tough day of work in a factory. As writer Alain de Botton observes in The Architecture of Happiness, “At the end of a shift in the plant, to be further reminded of the dynamism of modern industry was not a pressing psychological priority.”1
The families who occupied Le Corbusier’s homes defied his designs in a simple and practical fashion. They added old-fashioned shutters and windows; they erected pitched roofs over the flat ones; they put up flowery wallpaper over the uninterrupted monochrome walls, and marked out little blocks of garden with wooden picket fences.
Their gardens were decorated with gnomes.
• • •
Steve Jobs, one suspects, would not have approved of the gnomes. Jobs, like Le Corbusier, loved simplicity and clean modern lines—although unlike Le Corbusier, Steve Jobs had a gift for making things that people loved.
Jobs is most famous for his work on computers, phones, and tablets, galvanizing outstanding designers for three decades to produce some of the most beautifully crafted technology in the world. But Jobs also designed a building—the headquarters of Pixar, the animation studio that produced films from Toy Story to Inside Out. Jobs was Pixar’s majority shareholder.2
There can be no doubt that Steve Jobs was just as committed to the ideals of beauty as Le Corbusier had been, and he could be just as controlling. One of the saddest and most eloquent anecdotes in Walter Isaacson’s biography of Jobs describes him, semiconscious after receiving a liver transplant, ripping off his oxygen mask because it was ugly, and demanding that the medical team bring him five alternative designs so that he could pick the best.3
In happier days Jobs focused this fierce love for design on Pixar’s headquarters. In the wake of the smash hit Toy Story 2, he commanded a bigger budget than Le Corbusier. There was no need to save money using thin concrete walls. The Steve Jobs Building, named after Jobs’s death, is crafted of steel and glass, wood and brick, and Jobs obsessed over every detail. (It is easy to forget that the building also had an architect: Peter Bohlin, who designed the Apple stores.) Huge, beautiful steels were chosen after Jobs had pored over samples from across the country and selected a particular steel mill in Arkansas. He insisted that the girders were bolted rather than welded.4
And like all good designers, Steve Jobs cared about function as well as form.
“Steve had this firm belief that the right kind of building can do great things for a culture,” says Ed Catmull, Pixar’s president. “Steve wanted the building to support our work by enhancing our ability to collaborate.”5
Jobs had become fascinated by the idea of serendipitous interactions. How to make sure that everyone mingled together? He hit upon a plan: Pixar’s headquarters should have just a single pair of large restrooms off the main lobby. People would make new connections or revive old ones, because everybody would have to head to the lobby, brought together by a shared human need to urinate.6
Le Corbusier and Jobs were cultural giants, two of the most influential tastemakers of the last hundred years. But while they were unique, their love of tidy, minimalist spaces is common in the corporate world.
The “5S” system of management—Sort, Straighten, Shine, Standardize, and Sustain—has long stood for efficiency through tidiness and uniformity. The 5S system began in precision manufacturing spaces; clutter was discouraged because it might cause errors or delays, as were distracting personal effects. But 5S has somehow begun to bleed from car assembly lines and operating theaters and semiconductor manufacturing plants, where it might make sense, to the office cubicle, where it does not. Management gurus sing the praises of the “lean office.”
As The Wall Street Journal explained back in 2008, adherents of 5S at organizations such as Kyocera “are patrolling to make sure that workers don’t, for example, put knickknacks on file cabinets.” When the Journal went to Kyocera’s San Diego offices and followed the “5S inspector” around (he was a gently spoken middle manager by the name of Dan Brown), he was armed with a checklist and a digital camera as he cajoled his colleagues into removing illicit hooks from cubicle walls. He refused to be deflected by a clutterbug who had swept his junk into boxes and stored them in a closet. The closet was opened, the boxes inspected and photographed.7
This might seem intrusive, but as a management consultant breezily explained, “If managers clearly explain why they’re doing something, I think most people will understand the rationale.”
Perhaps. But then what is the rationale? The closest the Journal got to an explanation was that “to impress visitors, the company wants everything to be clean and neat.” This is why “sweaters can’t hang on the backs of chairs, personal items can’t be stowed beneath desks and the only decorations allowed on cabinets are official company plaques or certificates.”
Clear enough: things must look tidy to put on a good show for visitors. But will visitors really be won over by cabinets decorated with Kyocera’s own self-congratulatory plaudits? And if the aim is to impress visitors, why was Dan Brown earnestly examining the inside of filing cabinets and closets?
Le Corbusier’s vision was revolutionary; Steve Jobs’s was exacting; Kyocera’s was petty. But in all three cases one can see a very simple mistake being made. Each had failed to realize that what makes a space comfortable and pleasant—and, to turn to the concerns of modern business, inspiring and productive—is not a sleek shell or a tastefully designed interior. Indeed, it may have very little to do with how a building looks at all.
• • •
In 2010, two psychologists at the University of Exeter, Alex Haslam and Craig Knight, set up simple office spaces. Some were in a psychology lab, and some in a commercial office. Haslam and Knight recruited experimental subjects to spend an hour on administrative tasks such as checking documents. The idea was to see how the office environment affected how much people got done, and how they felt about it.
There were four different office layouts. The first was the lean office, a clean and spartan space with a bare desk, swivel chair, pencil, and paper. It reflected a Corbusian love of undecorated functionality, but the name “lean” was also inspired by the Japanese ideas of 5S that companies such as Kyocera had been pushing. It was Sorted, Straightened, Shined, and Standardized—although whether it could be Sustained was another question. It quickly became clear that the sheer tidiness of the space felt oppressive. “It just felt like a show space with nothing out of place,” commented one participant, adding, “You couldn’t relax in it.” Perhaps that is what proponents of office neatness intend.8
The second office layout enriched the lean office with some decorative elements. Large prints showing close-up photographs of plants hung on the wall. (One of the experimenters, Craig Knight, told me they reminded him of Georgia O’Keeffe paintings.) There were several potted plants, too. It may surprise modernists and fans of the 5S system to hear that workers got more done in the enriched office, and more accurately, while feeling better about their experience. It will not surprise anyone else.
The final two office layouts used the same components as the enriched office. Visually, they seemed much the same. In fact there was no visible difference at all between the pleasing and productive enriched office, and a space that was despised by all those who had the bad luck to have to work there.
The distinction in both cases was not how the office looked, but who got to decide its appearance. The most successful office space was called the empowered office. Like the enriched office it offered the same tasteful prints and the same little shrubs, but participants were invited to spend some time arranging those decorations however they saw fit. They could even have asked for them to be removed entirely, perfectly mimicking the lean space, if they wanted. The empowered office could be lean, or enriched, or something else; the point was that the person working in the office had the choice.
To produce the hated environment, the experimenters followed the same procedure, inviting participants to take time to arrange the prints and the plants however they wished. Once that had been done, the experimenter returned and began rearranging everything until the office precisely matched the enriched setting. The scientists called this condition the disempowered office, though that may be too mild a term. “I wanted to hit you,” one participant told the experimenter later, after the trial had been explained.
The empowered office was a great success—people got 30 percent more done there than in the lean office, and about 15 percent more than in the enriched office. These are large effects; three people in empowered offices achieved almost as much as four people in lean offices. The enriched office was a modest success, but the disempowered office, which offered exactly the same physical amenities, produced low productivity and low morale.
Haslam and Knight asked their participants a variety of questions about how much they rated the office they had been working in. They loved the empowered office and hated the lean and disempowered ones, complaining of being bored or even of physical discomfort such as feeling too hot. And their feelings of despair were all-embracing: if they disliked the office space, they also disliked the company that was hosting it, and disliked the task they were doing in it.
The physical environment certainly mattered, and—contrary to what Kyocera or Le Corbusier might believe—decorations such as pictures and plants tended to make workers happier and more productive. But there was much more to the environment than its design—equally important was who had designed it. The best option was to let workers design their own space. (At that point, it made no difference whether the space they created for themselves was sparse or enriched—presumably because each worker chose a space that worked for her.) The very worst was to give them the promise of autonomy, and then whisk it away.
But who would do such a thing? Let’s see how things are going over at Kyocera:
Mr. Scovie proudly showed Mr. Brown his tidy desk, which back in June could barely be seen for all the clutter, he admits. When asked for a peek inside his drawers, Mr. Scovie tried to steer the conversation to the placement of a desk blocking his access to some filing cabinets. Pressed, Mr. Scovie reluctantly agreed to open his drawers, one of which he warned was “really nasty.”9
The humiliation is palpable. The description evokes a parent nagging a child to tidy her room, or airport security patting down a suspected hijacker. One perfectly competent employee is being harassed by another perfectly competent employee to satisfy the pointless demands of a company rule book.
Haslam and Knight carried out the most explicit test of the importance of giving workers freedom to control their workspace, but other researchers have also pointed in that direction. In one study, NASA sent marine biologists to work for weeks on end in a tiny undersea lab—a truly tough environment, but the biologists loved it. However, they preferred to cook their own basic food from tins rather than eat elaborate food that had been prepared for them in advance.10 Robert Sommer, a psychologist at the University of California, Davis, spent years comparing “hard” and “soft” architectural spaces—those that people couldn’t change, and those that they could. Examples of “hard spaces” include those where the windows don’t open, the lights or air-conditioning cannot be changed, or the chairs are bolted to the ground. The quintessential hard space is a prison—but these prison-like features have spread to schools, public spaces, and the office. Sommer repeatedly found that apparently trivial freedoms, such as the right to paint your own wall, help people define personal space, and make people happier and more productive.11
Unfortunately, 5S enthusiasts at the Virginia Mason Medical Center in Seattle didn’t get the memo. Doctors and nurses there were in the habit of hanging a stethoscope on a hook, but management came up with a tidier solution: a drawer marked “stethoscope.” The medical staff kept on hanging the stethoscope on the hook. What to do? “Eventually,” said a supervisor, “we had to remove the hook.”12
The tidiness craze is global. At Her Majesty’s Revenue & Customs, the UK’s tax collection agency, staff were instructed in late 2006 to remove family photographs and souvenirs from their desks.13 At BHP Billiton, a vast mining company based in Australia, staff were ordered to maintain a clear desk as defined in an eleven-page instruction manual:
“Clear desk means that at the end of each day the only items remaining will be monitor(s), keyboard, mouse, mouse pad, telephone handset and headset, one A5 photo frame and ergonomic equipment (ie footstool, gel wrist pad etc).”14
If you wish to display an award, that’s okay—but only if you remove the A5 photograph. No plants allowed. And don’t think about ignoring the rules: “Facilities Management will consult with team managers about lapses.” All these anal-retentive rules are justified with much the same circularity as at Kyocera: clear desk policies are useful because they will keep things tidy, “to create and maintain a workplace that is clean, organised, and professional.”
It is one thing to sharpen and straighten all the pencils on one’s own desk, metaphorically or otherwise. To order someone else to sharpen and straighten the pencils on their own desk displays a curious value system in which superficial neatness is worth the price of deep resentment.
• • •
Use a spiral slide to get to the lobby! Have a team meeting in a pod shaped like a clown car! Enjoy risqué art installations! The business world is capable of embracing more than one fad for office design. While Kyocera, BHP Billiton, and the UK’s Revenue & Customs agency apparently favor the lab-rat chic of minimalist desks and bare cubicle walls, trendier companies prefer crazy materials and bright colors. And no company is or ever will be as with-it as the advertising agency Chiat/Day, creators of “1984,” the advertisement that launched the Apple Mac.
Amid gushing newspaper write-ups about the “workforce of the future,” Jay Chiat announced a radical plan to sweep away cubicles and offices and even desks. Armed with the best mobile technology available—in 1993—Chiat/Day employees would roam free in open spaces, winning sales and creating great ads wherever they wished. What’s more, those spaces would be playful, zany, and stylish. The agency’s new Los Angeles offices, designed by Frank Gehry and Claes Oldenburg, boasted a four-story sculpture of a pair of binoculars, and curvaceous two-seater pods from fairground rides were installed with the hope that people would sit together in them and think creative thoughts. The New York office, designed by Gaetano Pesce, had a mural of a vast red pair of lips, and a luminous multicolored floor with hieroglyphs all over it. Pesce had a boyish sense of humor: the floor in front of the men’s room had an illustration of a man urinating; his chairs had springs instead of feet, and would wobble and tip back so that when a woman sat on them, her colleagues could admire the view up her skirt; his conference tables were made of a silicone resin that would grab and hold important papers during meetings, with hilarious results. (It is not clear how many of these jokes were intentional.) The architect Frank Duffy, while admiring the daring of the project, comments, “Perhaps its gravest weakness is that it is a place where ‘play’ is enforced on everyone, all the time.”15
Jay Chiat saw himself as a revolutionary figure, a man chosen by destiny to tear down the conventions of office life. And, in some ways, his plans did prove visionary. Hiring Gehry showed impeccable taste; this was several years before the Guggenheim Bilbao would make Gehry the most famous architect on the planet. The bright colors, different architectural zones, clusters of couches interspersed with large tables, and use of mobile technology to free workers have all been widely copied. Chiat’s desire to make the office more like a university campus—“The idea is, you go to lectures, gather information, but you do your work wherever you like”—was also ahead of its time. Microsoft and Google each now refer to their corporate headquarters as a campus.
Overall, however, Chiat’s farsighted experiment was not a success. The results—as described in a superb article by Warren Berger—were “petty turf wars, kindergarten-variety subterfuge, incessant griping, management bullying, employee insurrections, internal chaos, and plummeting productivity. Worst of all, there was no damn place to sit.”16
Part of the problem was Chiat’s familiarly Kyocera-esque impatience with personalized workspaces. Chiat’s homes were decorated with modernist art rather than family photographs or souvenirs. He spent much of his time on airplanes, in any case. The idea that someone might want a desk cluttered with knickknacks was puzzling to him. Chiat dismissively commented that his staff could be given tiny lockers as a place to “put their dog pictures, or whatever.”
Chiat’s tidy-minded sensibilities went further than dog pictures: he tried to insist on an entirely digital, paperless office. If he saw storyboards taped on walls, or proofs of advertising posters, he’d demand that they be removed. What part of “paperless” didn’t people understand? Unfortunately, back in 1993, cell phones were temperamental and laptops were clunky, and both were expensive. Rather than own these essential tools, staff would sign them out each morning and return them to a concierge when they went home at night. The agency hadn’t bought enough for everybody, and this proved to be a false economy. Ill-tempered queues formed like breadlines each morning at the concierge desk. Staff who lived near the office would show up at six a.m., sign out a precious computer and phone, hide them somewhere, and then go back to bed for a couple of hours. Senior staff would order their assistants to rise early and secure their kit. Civil war broke out as each department claimed its work had priority over the others. Who would rely on digital documents if they couldn’t count on access to a computer?
With no personal workspace to call their own—Chiat would personally prowl the open-plan space, remembering where people had sat the day before and haranguing them not to “nest” in the same spot—Chiat’s employees had to improvise ways to keep hold of the paper copies of their contracts and storyboards and concept art. The personal lockers, intended for their dog pictures, were no good—they were too small to fit a standard binder. Some staff stashed binders in heaps in the corner. Others used their cars as storage, heading to the parking lot to file and retrieve important documents. One employee journeyed across the open prairie of the office with a little red wagon piled with her binders and papers. Not long after this story was reported, the Dilbert cartoon featured Wally pushing his work around in a shopping cart, tagging the office with graffiti and pondering joining a gang to give him a sense of identity.17
Playful offices such as Chiat/Day are often presented as radically different from traditional offices or cubicles. It is true that they reflect a different aesthetic, but Craig Knight—one of the scientists behind the research on workplace “enrichment” and “disempowerment”—points out that if the playfulness is inescapable, the philosophy ends up being much the same: management is the business of managers, design is the business of designers, and office serfs should simply accept what has been assigned to them. The tales of Chiat devoting some of his own scarce time to enforcing the rules is reminiscent of the researcher in Knight’s “disempowerment” setup driving an experimental subject to such rage that she wanted to hit him.
Chiat’s gaudy, postmodern tastes in architecture may be a world away from Le Corbusier’s simple modernist sensibilities, yet one similarity with Le Corbusier is striking: both men were unafraid to dream big dreams on behalf of other people. As the boss, Chiat insisted that his was the only vision that mattered. As one of Chiat’s top lieutenants, Bob Kuperman, summarized the matter: “Jay didn’t listen to anybody, he just did it.”
The story of Chiat/Day has a happy ending, after a fashion. Though the office makeover was dysfunctional, it looked fabulous in photo spreads, and design magazines went wild for it. Chiat/Day even started hosting paid tours of their office space. Having wrenched the spotlight onto his agency, Jay Chiat cashed in by selling his stake in the firm to Omnicom. Before long, the individual workspaces, paperwork, and dog pictures were back.
• • •
While Chiat/Day’s Gehry-designed headquarters were built to symbolize creativity, truly creative spaces often look very different. If you ask the veterans of MIT what a creative space looks like, one building comes to symbolize all that’s best at the university. This building, which was demolished a few years ago, didn’t even have a proper name—it was known only as Building 20—and it could hardly have been more different from the kind of structure that star architects put together.18
Building 20 was, quite literally, designed in an afternoon. In the spring of 1943, Don Whiston, a young architect and former MIT student, received a call from the university asking for preliminary plans and building specs for a 200,000-square-foot building by the end of the day. Whiston delivered. So did the builders who erected this squat, ugly, sprawling structure in plywood, cinder block, and asbestos with remarkable speed.
“I watched them put up Building 20,” one MIT professor recalled over half a century later.19 “You know those time-lapse photographs they make of construction of skyscrapers? Well, Building 20 went up like that—almost like that—in real time.”
There was a war on, after all. Building 20 was thrown together to house the Radiation Laboratory, or Rad Lab, a secret project that turned the primitive radars of the day into essential weapons of war. It was a vast research effort, and arguably more significant than the more celebrated Manhattan Project.
Unsurprisingly, given the hasty design and construction, Building 20 was a spartan, uncomfortable place to work. It was too hot in summer and too cold in winter. It was filthy with the dust of the just-finished building work. From the outside, it was a long, crude, three-story building, a cross between an army barracks and a garden shed. From the inside, it was a confusing, badly signposted labyrinth. And it was a firetrap: permission to build it was granted only on condition that it was pulled down within six months of the end of the war.
For two short years, Building 20 housed an astonishing cluster of researchers: one-fifth of the physicists in the United States passed through Rad Lab, which absorbed $2 billion of military funding—only $848,513 of which was spent on Building 20 itself. Nine Nobel Prize winners emerged from the Radiation Laboratory, and the technology they and their colleagues developed was an unqualified success: radars to allow planes to find U-boats; radars to provide early warning of V-1 attacks on London; radars to allow planes to land blind; radars to guide bombing raids; and more. The radar systems developed in Building 20 made previous technologies obsolete. The saying went that the atomic bomb may have ended the war, but it was radar that won it.
Given the temporary nature both of the Rad Lab wartime push and the building that housed it, you might imagine that Building 20’s triumphs were fleeting—a glorious moment in the history of applied science. And you might also assume that those triumphs took place despite the crude design and hasty construction of Building 20, not because of it. The authorities at MIT would have agreed with you. As promised, preparations were made to pull down the ugly, dysfunctional, and unsafe structure. The radio towers on the roof were removed; Rad Lab’s offices were dismantled. Building 20 would be replaced, in due course, with something more carefully designed and, well, a bit tidier.
Then came a stay of execution. The GI Bill of 1944 subsidized veterans to go to college. MIT was suddenly swamped by new students and desperately short of space. Building 20 was spacious and ready to be pressed into use. And so a temporary structure designed in an afternoon began a final act that was to last for decades.
It was quite a swan song. While we might assume that Rad Lab, the fruit of an extraordinary war effort, would have succeeded anywhere, the flow of ideas from Building 20 simply didn’t stop. It was the birthplace of the world’s first commercial atomic clock. One of the earliest particle accelerators was also constructed there. The iconic stop-motion photographs of a bullet passing through an apple were taken in Building 20 by Harold Edgerton. It was home to MIT’s Tech Model Railroad Club, a wellspring of hacker culture—hacks being skillful, innovative, and, yes, messy improvisations in pursuit of no goal other than the sheer joy of it. (In the 1950s, the hackers ran the model railway using components from the telephone exchange system.) From Building 20 emerged Spacewar, the first arcade-style video game. Meanwhile, from his office in Building 20, Jerry Lettvin was working on one of the most influential papers in cognitive science, “What the Frog’s Eye Tells the Frog’s Brain.” Noam Chomsky and Morris Halle revolutionized linguistics in Building 20.
It was in Building 20 that Leo Beranek built one of the first anechoic chambers, a room that swallows sound waves. The composer John Cage visited Beranek’s chamber and realized that, even there, he could hear his own blood being pumped around his body. This realization that there would always be ambient noise prompted him to compose 4'33", his infamous four and a half minutes of silence.* Meanwhile, Beranek founded the company Bolt, Beranek and Newman, which moved from developing acoustic-processing computers to building some of the first Internet networks and inventing e-mail as we know it today. Leo Beranek wasn’t the only acoustic pioneer at Building 20: a young electrical engineer named Amar Bose, dissatisfied with a piece of hi-fi equipment he’d purchased, wandered down the corridor to the acoustics labs to see whether anything better could be created. There, he revolutionized the speaker, and established the Bose Corporation. DEC, the Digital Equipment Corporation, a major force in the pre-PC era of computing, was another significant technology company incubated in Building 20.
All this took place in an atmosphere of chaos and neglect—to the extent that a storeroom in Building 20 was taken over by a homeless botanist who haunted the corridors in the 1960s and 1970s. MIT tried to evict him, and lost the case. Surely a myth? And yet both Jerry Lettvin and Morris Halle testified to his existence. “He turned down a job at the Field Museum in Chicago in order to remain a phantom in Building 20,” Lettvin told The Boston Globe.
Building 20 was ugly and uncomfortable but its occupants loved it. MIT’s president during the 1970s, Jerome Wiesner, described it as “the best building in the place,” while Jerry Lettvin said it was “a building with a special spirit, a spirit that inspires creativity and the development of new ideas,” adding that it was the “womb of the Institute. It is kind of messy, but by God it is procreative!”
The question is, why?
• • •
When people praise Building 20, they often point to something that Steve Jobs himself would have admired: the building’s propensity to throw people together at random. This was as much by accident as anything; for instance, Building 20 had a baffling system of office numbering. If you wanted to find the Office of Naval Research—a descendant of Rad Lab—it was in Room 20E-226. But where exactly was that? The 20 of 20E-226 was clear enough: it referred to Building 20. The “E” referred to E wing, which was parallel to and sandwiched between A wing on one side and D wing on the other. C wing was farther along. B wing was the spine that linked all the other wings together. (In a more logical world, B should have been A, A should have been B, E should have been C, and C should have been E.) The 226 meant room 26 not on the second story but on the third, Building 20 being one of the few buildings in America to have unilaterally adopted the British system for labeling floors.
This absurdly inefficient way of organizing a building meant that people were constantly getting lost and wandering into places they didn’t intend to go. Better still, because Building 20 was low-rise and sprawling, when chance meetings occurred, they didn’t happen in elevators, the eternal home of the glib, tidy monologue we call the “elevator pitch.” They began in long corridors, where a genuine conversation could develop.
More important, the combination of people who could have those conversations was strange and wonderful. In the early 1950s, Building 20 contained departments that were wartime holdovers—nuclear science, flight control, the “Guided Missiles Program Office”—but also plastics research, the adhesives lab, the acoustics lab, the electronics lab, and even an outpost of the architecture department: a lighting design shop. Over the next ten years, MIT’s data processing group was added, along with the ice research lab, MIT Press, and the student hackers at the Model Railroad Club. Into the mix were stirred machine shops for the nuclear scientists and for the electronics research lab, the photography labs, a materials lab for the anthropologists, and solar car researchers who would use the building’s corridors as driveways and parking lots. The building even hosted a piano repair shop (a “computer free” zone, warned a sign on its wall) and the Reserve Officers’ Training Corps (ROTC) offices, right next to the office of the anti-establishment linguist Noam Chomsky.*
This unlikely mess made possible chance interactions among innovative researchers that paid such spectacular dividends. Who would have guessed that throwing the electrical engineers in with the Model Railway Club would result in hacking and video games? Or that the electronics specialists, the music department, and the acoustics lab would end up spawning technology pioneers such as the Bose Corporation and Bolt, Beranek and Newman?
Nobody would have guessed, and nobody tried to guess, either. The hodgepodge of Building 20 was the result of simple expedience and neglect. Where did MIT put disciplines that didn’t fit, researchers who had no clout, projects that made no money, student hobbyists, and anything and anyone else that just didn’t seem to matter? In the cheapest, nastiest space they could find. If Building 20 hadn’t been a mess, these strange collaborations might never have happened.
Another key element of Building 20’s success was that the space was easy to reconfigure. Its services—water, phones, electricity—were exposed, running along the corridor ceilings, supported by brackets. This was ugly but convenient. Researchers thought nothing of tapping into them directly for whatever experimental needs they had. Paul Penfield, a longtime occupant of Building 20, recalled: “You know that if you want to run a wire from one room to another, you don’t call Physical Plant, you don’t plunk down a thousand dollars to call an electrician and a carpenter, instead you get out a power drill or a screwdriver, and you jam it through the wall, and you string the wire, and you take care of things right away, and you do it in one afternoon, rather than waiting six months for a purchase order to come through.”
Can you imagine what the clean-desk brigade at Kyocera would have made of that?
In some ways, a modern office designer would approve of Building 20: contemporary office buildings today are routinely designed so that partition walls can be added or removed to make whatever space is necessary. But few modern offices boast the extreme reconfigurability of Building 20: when the atomic clock was being developed by a team led by Jerrold Zacharias, the group simply removed a couple of floors from their laboratory to accommodate it.
And Building 20’s true advantage wasn’t so much that it was reconfigurable by design, but that the building’s inhabitants felt confident that they had the authority (if only by default) to make changes, even messy changes. It was that it was so cheap and ugly that in the words of Stewart Brand, author of How Buildings Learn, “Nobody cares what you do in there.”
Unlike at Kyocera or Chiat/Day, Building 20’s inhabitants were in control. Heather Lechtman, professor of material science and archaeology, told Brand: “We feel our space is really ours. We designed it, we run it.” After Jerome Wiesner became president of MIT in 1971, he kept a secret office tucked away in Building 20. Why? Because “nobody complained when you nailed something to a door.”
A university or corporate research center can and should create interdisciplinary spaces in which well-established teams seek common problems to work on. But the anarchy of Building 20 went way beyond what any official effort at cross-cultural collaboration is ever likely to tolerate. It would be a brave CEO who’d play host to model railway enthusiasts and homeless botanists.
When Building 20 was at last demolished in 1998, MIT held what could only be described as a wake. Engineering professor Paul Penfield organized the commemoration “to help each other through the grieving process.”
Building 20 was replaced by the ultimate architectural status symbol—a building designed, like the Chiat/Day headquarters, by Frank Gehry. Gehry’s Stata Center, which opened in 2004, is designed as a symbol of creativity. Imagine one of Dr. Seuss’s architectural fantasies bursting fully formed onto the campus of the Massachusetts Institute of Technology, and you begin to get the picture. The architecture critics loved the center’s messy appearance. In The Boston Globe, Robert Campbell gushed, “The Stata’s appearance is a metaphor for the freedom, daring, and creativity of the research that’s supposed to occur inside it.”20
But being a metaphor for freedom, daring, and creativity is not the same as actually being conducive to it. The cheap and cheerful Building 20 was a proper mess. But its computer-modeled replacement, painstakingly constructed at the cost of $300 million, was more of a tidy-minded person’s idea of what a mess looks like. The Stata Center soon developed problems, including a tendency for snow to build up on the building’s wacky windows and then cause damage by sliding off. MIT administrators were sufficiently irked to take Gehry, and the building contractors, to court.21
But the real problem with the structure wasn’t a buildup of snow. Shortly before the Stata Center was due to be occupied, Wired magazine caught up with one of the MIT greats, the computer guru Gerald Sussman, and asked him what he thought of the new building.
With a thin smile of contempt, he said all that needed to be said: “I didn’t ask for it.”22
• • •
It is still fashionable, especially in creative industries, to follow the Chiat/Day example and install playground equipment and quirky decorations in the office. For the past decade, the most-written-about corporate headquarters has been Google’s “Googleplex” in Mountain View, California, with its Ping-Pong tables and slides. The media remain as giddy as ever about unusual corporate environments.
But it is easy to fall victim to a logical error here. Google wasn’t successful because it built the toy-filled Googleplex. It built the toy-filled Googleplex after it had already become successful, just as Chiat/Day commissioned Frank Gehry after being the most admired advertising agency of the 1980s.
A closer look at the history of Google’s headquarters suggests a closer resemblance to MIT’s Building 20 than to Chiat/Day’s hot-lipped theme park. For the first two years of Google’s gestation, while Sergey Brin and Larry Page were making some of the foundational breakthroughs, there were no headquarters at all: Brin and Page were studying at Stanford University.23
In September 1998, Google moved to the clichéd start-up location: a garage. They rented some rooms, too, in a house on Santa Margarita Street in Menlo Park. One room contained Sergey, Larry, and two other engineers. The garage itself was packed with servers. Desks were the simplest possible design: a door placed horizontally across a pair of sawhorses. Nothing could be cruder or easier to put together and take apart, or easier to hack about. One day the house’s owner, Susan Wojcicki, was expecting delivery of a refrigerator. She returned home to find that the Googlers had commandeered it, moved it to their part of the house, and filled it with drinks and snacks. This was the typical “anything goes” scavenger mentality of early Google. Wojcicki didn’t seem to mind—she later joined the company and ended up running YouTube.24
In any case, by the spring of 1999, Google had moved again, to an office over a bicycle shop. The desks, again, were doors on sawhorses, and the engineers had decided to bring up a Ping-Pong table. Larry and Sergey left red and blue inflatable gym balls around, less as an aesthetic statement and more because they liked to work out with them.25 Google soon moved again, to an office-park facility in Mountain View. This office space later became known as the NullPlex, the space before the Googleplex itself. But again, the space was crude—a “mishmash,” said the facilities manager, George Salah. A “mongrel style,” said Steven Levy, journalist and unofficial historian of Google. One of the first high-priority projects to be tackled in the new space was the problem of how to make searches more responsive to the latest news. Google set up a war room, and again, it was a straightforward, unassuming space. Half a dozen engineers grabbed a conference room, set up their computers the way they wanted, and got to work.
Then there was the time that one Google engineer decided he didn’t like the wall of his office. When Google’s facilities manager came in the next morning, he was surprised to discover that the engineer and his colleagues had knocked it down. But he didn’t complain. Neither did he complain when the engineer later changed his mind and decided he’d like to put the wall back again; instead, he mused that the process had “made it a more Googley environment.”
Any veteran of MIT’s Building 20 would recognize the thought process. And when the suit-and-tie executive Eric Schmidt joined Google as the new boss in 2001, he reassured Salah, “Don’t change a thing. Make sure it looks like a dorm room.”26
“No matter what happened,” writes Steven Levy, “engineers would have the run of the place.”27
• • •
The offices at Chiat/Day may have looked superficially different from the offices at Kyocera, but they were managed with fundamentally the same tidy-minded aesthetic: This place should look the way the boss wants it to look. Google’s offices, like Building 20 at MIT, have been managed very differently: It doesn’t matter how this place looks. The denizens of Building 20 had tremendous power over their environment, even by the standards of tenured academics. This wasn’t because of their high status. If anything the opposite was true: the higher-status academics were placed in high-status spaces that were too beautiful, expensive, and historically significant to tinker with. (No wonder Jerome Wiesner, MIT’s president, felt trapped by the magisterial perfection of his office, and kept a messy bolt-hole in Building 20.) The lower-status academics were placed in low-status spaces, and that meant they could do whatever they wished. Nobody cared.
People flourish when they control their own space. But if it causes so much damage to impose a rigid aesthetic on workers, why are bosses so keen to keep everything shipshape? It seems as though actively encouraging staff to take control is far scarcer than the benign neglect of Building 20. Why is creativity something that happens only when the boss isn’t looking?
Some clues come from the remarkable career of Robert Propst. Propst was a sculptor, painter, art teacher, and inventor of devices as varied as a vertical timber harvester and a machine-readable livestock tag. A trained chemical engineer, he’d spent the Second World War managing beachhead logistics in the South Pacific. In 1958, he was hired by the Herman Miller company, a manufacturer of office furniture. Herman Miller’s managers thought Propst was a genius.28
He was certainly an independent spirit: he stayed in Ann Arbor, Michigan, 150 miles away from Herman Miller’s headquarters in Zeeland, later persuading the company to set up a research division there to accommodate him. Propst continued his inventions, and in office furniture circles he is famous for creating the “Action Office II.” Launched in 1968, this modular system of parts could be clipped together at funky angles, allowing office workers to craft an environment suited to their own needs, like children tinkering with Legos. Propst imagined a free-flowing office environment with an egalitarian ideal. Good-bye to a boss’s desk that looked like a mahogany altar, and to the interchangeable serfs of the typing pool; hello to autonomy, to workers who could “behave like a manager at every level.”
But then America’s office managers made a few modifications. Propst’s original dividers clipped together at an angle of no less than 120 degrees, providing a fanned-out space on which to pin working documents. Managers demanded dividers with 90-degree angles, allowing micro-office after micro-office to snap together in regimented lines. They loved the fact that Action Office II’s partitions didn’t count as walls, but as furniture, which earned a more generous tax treatment. And so the office cubicle was born.
Robert Propst had been committed to the idea of the empowered worker. He knew that good design meant giving workers control over their own environment. But he was helpless in the face of corporate bosses more interested in saving money than in his progressive design ideals. Propst was left to condemn the perversion of his ideas as “monolithic insanity,” “hellholes,” “egg-carton geometry,” and “barren, rat-hole places.”
When the father of the cube farm died in 2000, workplace cubicles had never been more ubiquitous. Empowerment is all very well, but never underestimate what managers will inflict on workers in an effort to avoid tax and minimize rents.
If cubicles are attractive because they are cheap, Propst understood that other forces are also at play. Managers could be tidy-minded simply because tidiness seems like the right and proper way to be. “We all have this desire for formal order,” Propst had written in 1968. “The only problem is that it conflicts severely with the more organic kind of spatial order human interchange uses best.”29
So far, the desire for formal order is winning. We like tidiness to the point of fetishizing it; we find clutter and irregularity disturbing and don’t notice when it is doing us good.
T George Harris, a veteran journalist and editor of Psychology Today, put his finger on the problem back in 1977:
The office is a highly personal tool shop, often the home of the soul . . . this fact may sound simple, but it eludes most architects, office designers, and thousands of regulations writers in hundreds of giant corporations. They have a mania for uniformity, in space as in furniture, and a horror over how the messy side of human nature clutters up an office landscape that would otherwise be as tidy as a national cemetery.30
Harris details some of the fussier corporate policies of the day—at CBS, where “a producer could not look at film slides in his office without first nursing two interoffice memos up through the facilities hierarchy—one memo to dim overhead lights, the other to move a chair.” Harris sounds like he’s joking, but he isn’t. Another supposedly creative business, the publisher McGraw-Hill, “set up a Deviation Committee to rule on any request to make any change in the physical setup specified in a positive blizzard of advance memos.”
Consider the following piece of well-worn advice from fifty years ago, in The Business Etiquette Handbook:
Avoid over-decorating your desk or area. When your desk, shelves, and wall space are covered with mementoes, photographs, trophies, humorous mottoes, and other decorative effects, you are probably not beautifying the office; rather you may be giving it a jumbled, untidy look . . . The proper atmosphere for a business office is one of neatness and efficiency, not hominess.31
The unexamined assumption is that this jumbled untidiness is bad, and that if the office looks streamlined then it will also be more productive. But the evidence suggests the opposite. Yet fifty years later, bureaucrats at Kyocera and BHP Billiton are continuing with these tidy traditions.
Is there any evidence that neat environments really help? The longer T George Harris chased credible research into the impact of “good design” on employee productivity, the more elusive that research seemed. “People suddenly put into ‘good design’ did not seem to wake up and love it,” he wrote. What they loved instead was control over the space in which they had to live or work.
And that control typically leads to mess. The psychologist Craig Knight admits that a space that workers design for themselves will almost always look rather ugly. “It doesn’t look as good as something a designer would have chosen, and it never will.”32
The management theorist A. K. Korman vividly recalls visiting one factory where the mess had been embraced:
I was assaulted with a kaleidoscope of orange, blue, pink, yellow, red and multi-colored machines. My host laughed at the expression on my face and then went on to tell me that the management of the company had told the workers they could paint the machines any color they wanted and the company would furnish the paint if they furnished the manpower. The result was a very unusual looking factory to me, although it was a pleasing work environment to those who worked there every day.33
From the vantage point of a nice corner office, someone else’s messy desk is an eyesore. The clutter is visible, but the resulting sense of empowerment is not. For the senior manager, the lesson is simple: Resist the urge to tidy up. Leave the mess—and your workers—alone.
• • •
When Steve Jobs got an important idea in his head, it wasn’t easy to dissuade him. His plan to impose a single pair of serendipity-inducing über-bathrooms on Pixar seemed to be a very important idea indeed. Random connections mattered, thought Jobs, and he was right. And what better way to promote random connections than by forcing people to go through the lobby several times a day, at intervals governed by the call of nature?
“He felt that very, very strongly,” says Pam Kerwin, Pixar’s general manager.
So Jobs explained his idea to Pixar’s staff at an off-site meeting to discuss the plans, and the staff didn’t like it at all. As Kerwin recalls, “One pregnant woman said she shouldn’t be forced to walk for ten minutes just to go to the bathroom, and that led to a big fight.”
John Lasseter, Pixar’s creative director, stood up for the pregnant woman. Jobs was frustrated. People just didn’t understand the vision. They didn’t get it.
But then Jobs did something extraordinary and out of character. He backed down. The Steve Jobs Building contains not one, but four pairs of bathrooms.34
There is still plenty of opportunity for serendipity, thanks to an atrium that focuses activity, with the main doors of the building, a café, a game area, the mailboxes, three theaters, conference rooms, and screening rooms all spilling into it. John Lasseter says that Jobs’s basic instincts had been correct: “I kept running into people I hadn’t seen for months. I’ve never seen a building that promoted collaboration and creativity as well as this one.”
Pixar’s boss Ed Catmull agreed. “People encountered each other all day long, inadvertently, which meant a better flow of communication and increased the possibility of chance encounters. You felt the energy in the building.”35
No doubt that is true. But something else matters just as much as serendipity: autonomy. Junior staff were able to stand up to Steve Jobs, the owner, the legend, the control freak’s control freak—and to get their own way about something that mattered to them. That was more important than all the riveted steel and elegant brickwork Pixar’s success could buy.
One day, Catmull found himself reflecting on a gorgeous table in Pixar’s main conference room. It was long and elegant, chosen by one of Jobs’s favorite designers. It reminded Catmull of a comedy sketch in which a wealthy couple sat down for a dysfunctional dinner, the man at one end, the woman at the other, a vast candelabra in the center, and no possibility of having a conversation. Catmull realized that although the table was beautiful, its long, thin design was dysfunctional when it came to the open, egalitarian discussions Pixar claimed to value.36
At meetings, thirty people faced off against each other in two lines, and if they were to hear everyone, the high-status people had to cluster around the middle. That would be Catmull, Lasseter, and the director and producer of the film in question. There was an informal but perfectly real hierarchy, as lesser people had to sit farther and farther away from the conversation and found it hard to get a word in. The hierarchy was gradually formalized with the introduction of place cards.
Catmull admits he didn’t notice for over a decade—after all, the conversation seemed fine to him. Eventually he realized the problem after having a much better discussion sitting around a less elegant table in a different room. In a couple of days, the beautiful table in the main conference room was gone. There is more to creativity than elegance.*
Staff autonomy continues to flourish at Pixar. The most famous example is a concealed room that can be reached only through a crawlway and which was originally designed merely to provide access to the air-conditioning valves. Once a Pixar animator discovered the secret panel into the space, he installed Christmas lights, lava lamps, animal print furnishings, a cocktail table, a bar, and napkins printed up with the logo “The Love Lounge.” When Steve Jobs found out about the Love Lounge, he loved it.
“The animators who work here are free to—no, encouraged to—decorate their work spaces in whatever style they wish,” explains Catmull. “They spend their days inside pink dollhouses whose ceilings are hung with miniature chandeliers, tiki huts made of real bamboo, and castles whose meticulously painted, fifteen-foot-high Styrofoam turrets appear to be carved from stone.”37
It sounds like a mess.